This proposal has two mutually supporting goals; to advance the design of multisite microstimulating arrays and microstimulation technology for their use in rehabilitation medicine and basic neuroscience, and to advance the development of cochlear nucleus auditory prostheses towards providing speech recognition that is at least equivalent to that of users of cochlear implants. Persons who lack a functional auditory nerve cannot benefit form cochlea implants, but some hearing can be restored by a prosthesis implanted in or on the cochlea nucleus. However, the devices now in clinical use do not restore hearing that is comparable to that of cochlea implants. They are mechanically sturdy and with their ground tips can be inserted into the brain with minimal trauma. In a cat model, we will evaluate the safety of 140 hours of microstimulation in the cat cochlear nucleus at a pulse rate of 500 pps, in order to determine the roles of stimulus pulse rate and stimulus charge density in stimulation-induced neuronal injury when the pulse rate is high (250 to 500 pps), and the electrodes' geometric surface areas is in the upper part of the range for microstimulation (2,000 to 4,000 m2). We will determine a combination of electrode geometric surface area and stimulus charge per phase that is not injurious to the neurons and other cell types close to the electrodes. Functional electrical stimulation in the central nervous system with penetrating microelectrodes has potential applications in clinical medicine and basic neural science, and high-rate stimulation can convey more temporal information and may elicit neuronal activity that more closely resemble naturally-occurring activity by minimizing locking of neuronal activity to the individual stimulus pulses, and may find other uses in clinical medicine, including treatment of movement disorders. In a study just completed, we found that encoding of amplitude modulation by microstimulation in the cochlear nucleus is improved by using a pulse rate of 500 pps. We will enhance our multisite silicon substrate microstimulation probes in order to increase their lifetime in vivo, with the goal of qualifying them for clinical use. Our devices have 5 independent electrode sites on each shank, allowing placement of a large number of electrodes into the target nucleus with the minimum tissue displacement and damage .The silicon shanks are mechanically sturdy and their ground tips can be inserted into the brain with minimal trauma. We will develop probes of this configuration that will meet at least the following performance standards after 1 year of soak in buffered saline at 39oC , and for which accelerated testing indicates that the standards will be met for at least 8 years (1) ; inter-channel crosstalk (channe interaction) between the electrodes on the same probe shank during controlled-current pulsing below 5% for all channels on the probe shank and (2) the leakage impedance of each channel to the saline bath greater than 1 Mgs . A multisite stimulating array that maintains these performance standards will retain essentially full functionality. We will optimize procedures for encoding the amplitude modulation (AM) of sound into an electrical stimulus that is applied in the ventral cochlear nucleus. Based on work completed in our laboratory on encoding of amplitude modulation by neurons in the inferior colliculus of the cat, this will require a high stimulus pulse rate (500 pps). This work also will serve as a test bed for the long-lived microstimulation arrays we will develop.